pipeline_tag, inference, widget, license, datasets, metrics, library_name, tags, model-index

pipeline_tag	inference	widget	license	datasets	metrics	library_name	tags	model-index
text-generation	true	text example_title group Question: Please write a function in Python that performs bubble sort.\n\nAnswer: Bubble sort Python	bigcode-openrail-m	bigcode/commitpackft bigcode/oasst-octopack	code_eval	transformers	code	name results OctoCoder task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalSynthesize Python name type value verified pass@1 pass@1 46.2 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalSynthesize JavaScript name type value verified pass@1 pass@1 39.2 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalSynthesize Java name type value verified pass@1 pass@1 38.2 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalSynthesize Go name type value verified pass@1 pass@1 30.4 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalSynthesize C++ name type value verified pass@1 pass@1 35.6 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalSynthesize Rust name type value verified pass@1 pass@1 23.4 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalSynthesize Average name type value verified pass@1 pass@1 35.5 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalFix Python name type value verified pass@1 pass@1 30.4 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalFix JavaScript name type value verified pass@1 pass@1 28.4 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalFix Java name type value verified pass@1 pass@1 30.6 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalFix Go name type value verified pass@1 pass@1 30.2 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalFix C++ name type value verified pass@1 pass@1 26.1 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalFix Rust name type value verified pass@1 pass@1 16.5 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalFix Average name type value verified pass@1 pass@1 27.0 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalExplain Python name type value verified pass@1 pass@1 35.1 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalExplain JavaScript name type value verified pass@1 pass@1 24.5 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalExplain Java name type value verified pass@1 pass@1 27.3 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalExplain Go name type value verified pass@1 pass@1 21.1 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalExplain C++ name type value verified pass@1 pass@1 24.1 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalExplain Rust name type value verified pass@1 pass@1 14.8 false task dataset metrics type text-generation type name bigcode/humanevalpack HumanEvalExplain Average name type value verified pass@1 pass@1 24.5 false

Table of Contents

1. Model Summary
2. Use
3. Training
4. Citation

Model Summary

OctoCoder is an instruction tuned model with 15.5B parameters created by finetuning StarCoder on CommitPackFT & OASST as described in the OctoPack paper.

• Repository: bigcode-project/octopack
• Paper: OctoPack: Instruction Tuning Code Large Language Models
• Languages: 80+ Programming languages
• OctoPack🐙🎒:

Data	CommitPack	4TB of GitHub commits across 350 programming languages
	CommitPackFT	Filtered version of CommitPack for high-quality commit messages that resemble instructions
Model	OctoCoder	StarCoder (16B parameters) instruction tuned on CommitPackFT + OASST
	OctoGeeX	CodeGeeX2 (6B parameters) instruction tuned on CommitPackFT + OASST
Evaluation	HumanEvalPack	Extension of OpenAI's HumanEval to cover 3 scenarios across 6 languages

Use

Intended use

The model follows instructions provided in the input. You should always preface your input with "Question: " and finish it with "Answer:", for example: "Question: Please write a function in Python that performs bubble sort.\n\nAnswer:"

Feel free to share your generations in the Community tab!

Generation

# pip install -q transformers
from transformers import AutoModelForCausalLM, AutoTokenizer

checkpoint = "bigcode/octocoder"
device = "cuda" # for GPU usage or "cpu" for CPU usage

tokenizer = AutoTokenizer.from_pretrained(checkpoint)
model = AutoModelForCausalLM.from_pretrained(checkpoint).to(device)

inputs = tokenizer.encode("Question: Please write a function in Python that performs bubble sort.\n\nAnswer:", return_tensors="pt").to(device)
outputs = model.generate(inputs)
print(tokenizer.decode(outputs[0]))

Training

Model

• Architecture: GPT-2 model with multi-query attention and Fill-in-the-Middle objective
• Steps: 250k pretraining & 30 instruction tuning
• Pretraining tokens: 1 trillion pretraining & 2M instruction tuning
• Precision: bfloat16

Hardware

• Pretraining:
  • GPUs: 512 Tesla A100
  • Training time: 24 days
• Instruction tuning:
  • GPUs: 8 Tesla A100
  • Training time: 4 hours

Software

• Orchestration: Megatron-LM/Transformers
• Neural networks: PyTorch

Citation

@article{muennighoff2023octopack,
      title={OctoPack: Instruction Tuning Code Large Language Models}, 
      author={Niklas Muennighoff and Qian Liu and Armel Zebaze and Qinkai Zheng and Binyuan Hui and Terry Yue Zhuo and Swayam Singh and Xiangru Tang and Leandro von Werra and Shayne Longpre},
      journal={arXiv preprint arXiv:2308.07124},
      year={2023}
}